03.26.2007
Subprime Derivatives
As if subprime mortgages aren’t scary enough, there are financial derivatives based on subprime loans. (Nouriel Roubini sounds worried. Macroblog is a calmer.) Certainly, there is a broad set of questions about the social benefits and costs of credit derivatives, something ex business-school professor and now Governor Randall Kroszner discusses here.
But we think this market data gives some useful information about where the smart money sees the risks, and in contrast to the default and delinquency numbers, gives a more forward-looking take on the problem.
This is the ABX.HE index, which is based on credit default swaps on different tranches of subprime mortgage-backed securities (MBS). Admittedly, unless you’re a financial markets junky it’s not at all clear what this index is saying, other than something called BBB is moving a lot more than the stuff called AAA. Let’s try to deconstruct what’s going on, keeping in mind that this is most definitely not investment advice.
First, the ABX.HE index is telling us something about credit default swaps (CDS). A CDS is like a derivative that gives you insurance. For example, a bank may wish to buy protection against default by RiskyCorp (perhaps because they’ve given RiskyCorp a loan). They do this by entering into a contract where they pay another firm (who is selling protection) a fixed amount, periodically, as long as RiskyCorp doesn’t default on its corporate bonds. (In general, the “credit event” might be something else, such as a major downgrade, missed payments, or so forth.) If RiskyCorp does default, the seller of protection makes a payment to the buyer of protection. This might be a cash payment equal to the value of the bond, it might be the bond itself, or potentially whatever the contracting parties agree to. We like to think of the “swap rate” or “swap spread” that the protection buyer must pay as an insurance premium.
Notice that the more likely RiskyCorp is to default, the higher the insurance premium, that is, the higher the swap spread, so this market can give us some idea of how risky some firms are. In a frictionless market, the swap spread should be comparable to the risk premium on one of RiskyCorp’s corporate bonds (corporate bond yield minus the comparable riskless yield). Furthermore, because the CDSs are more standardized and generally more liquid than corporate bonds, you can see why Federal Reserve Vice Chairman Donald Kohn states that “instead of looking to the bond market to measure default risk, we are increasingly turning to the market for credit default swaps” (the full text of his remarks to the Board Conference on Credit Risk and Credit Derivatives is well worth reading).
Credit default swaps eventually became based on other types of assets, such as mortgage-backed securities, whose payoff is derived from a pool of mortgages (such asset-based swaps became known as ABCDS, for obvious reasons). Likewise, there was no reason to restrict your CDS so that it protected you against default from only one firm, and although such “single-name” CDSs still make up the bulk of the market, “multiname” CDSs are growing in popularity.
Mortgage-backed securities offer several different levels of risk or tranches. Tranches are ways of slicing up the payment stream from homeowners to give different levels of risk, so roughly speaking, the tranches first in line for payments are less risky than those further down the line.
At long last: the ABX.HE is a series of five indexes that track CDSs based on tranches of mortgage-backed securities comprised of subprime mortgages and home equity loans. The tranches differ by their ratings, from AAA (best credit) to BBB-, (least good credit). See MarkiT, which produces the indexes for the real details. For an example of how indexes work, see here.
The prices of the riskier tranches started moving down in late 2006, but the real action started in late 2006, so a closer look is appropriate. There’s not a lot of movement among the top-quality tranches, the AAA and AA, but February (when New Century Financial and HSBC, the number-three and number-two suprime lenders, announced problems) was rough on the lower-rated indexes, with the BBB- dropping from 90.85 to 64.46. That’s a 29 percent drop in only one month. Since then, there’s been a rebound, up over 10 percent, but the market seems to be anticipating continuing large losses in the subprime market.
Of course, changes in the price of risk, in liquidity, even the volatility of other interest rates can materially affect these prices, so as always, it’s best to treat any inferences you may wish to draw with caution.
03.21.2007
What's Up with the Yield Curve?
The recent news on the housing market, stock price volatility, and the Yen-carry trade has perhaps overshadowed one persistent trend in the financial markets: The yield curve remains inverted. With the 10-year rate at 4.78 percent and the 3-month rate at 5.07 percent (those are constant maturity for the week ending March 16, from the H15 release for you quants out there), the spread stands at a negative 29 basis points, and indeed has been underwater since August.
Should anyone but the bond geeks really care? Of course, one can look for deep reasons behind the flat yield curve and consider what it says about global market forces, something better minds than ours have pondered: see here and here.
But usually, if the yield curve hits the headlines, people think it’s predicting something. The curve’s slope has achieved some notoriety as a simple forecaster of economic growth, the rule of thumb being that an inverted yield curve (short rates above long rates) indicates a recession in about a year. Yield curve inversions have preceded each of the last six recessions (as defined by the NBER). Very flat yield curves preceded the previous two, and there have been two notable false positives: an inversion in late 1966 and a very flat curve in late 1998. More generally, though, a flat curve indicates weak growth, and conversely, a steep curve indicates strong growth. One measure of slope, the spread between 10-year Treasury bonds and 3-month Treasury bills, bears out this relation, particularly when real GDP growth is lagged a year to line up growth with the spread that predicts it.
Projecting forward using past values of the spread and GDP growth suggests that real GDP will grow at about a 1.7 percent rate over the next year.
History notwithstanding, predictions from the yield curve show a lot more pessimism than most other forecasts, such as Blue Chip.
Even the most pessimistic bunch of the Blue Chippers look downright giddy compared to the yield curve.
Believe it or not, that’s the optimistic spin from the yield curve. Another popular approach foregoes predicting actual GDP growth and focuses on using the yield curve to predict a discrete event: whether or not the economy is in recession. Looking at that relationship, the expected chance of a recession in the next year is 46 percent, up a bit from last month’s value of 42 percent.
Macroblog has called this “the bears’ strongest suit.” On the other hand, the recent woes in the subprime mortgage industry are making pessimism a bit more fashionable these days.
Of course, it might not be advisable to take this number quite so literally, for two reasons. First, this probability is itself subject to error, as is the case with all statistical estimates.
Second, the yield curve’s predictive power rests on a correlation, (albeit one that’s been around for a while), and there is not an agreed upon theory to explain it (though some definite contenders are already in the ring: see here and here). Still, if the underlying determinants of the curve should change, the simple forecasting tool will prove a little too simple. Some researchers have postulated that those underlying determinants have changed, and are materially different from those that generated yield spreads during prior decades. Differences could arise from changes in international capital flows and inflation expectations, for example. And, to be honest, there are some heavyweights on that side of the issue (see here and here).
The bottom line is that yield curves contain important information for business cycle analysis, but, like other indicators, it should be interpreted with caution.
For more detail on these and other issues related to using the yield curve to predict recessions, see the Commentary "Does the Yield Curve Signal Recession?"